• OpenAI launches a limited beta test of clearly labeled ads in ChatGPT for a subset of free and Go tier users in the U.S., marking a strategic shift toward monetization amid scaling costs.
  • The company sets premium advertising terms with $200,000 minimum commitments and $60 CPM rates, comparable to high-end TV ads, while keeping Plus and Pro tiers ad-free.
  • This move follows CEO Sam Altman's past reservations about AI ads as "uniquely unsettling," now positioning them as a "last resort" to fund growth, with privacy assurances and contextual targeting.

A Strategic Pivot Toward Monetization

OpenAI has quietly begun rolling out a test for ads in ChatGPT today, targeting a subset of free and Go tier users in the U.S. as part of a limited beta initiative. According to people familiar with the matter, the ads appear at the bottom of responses, are clearly labeled, contextually relevant, and dismissible, with the company emphasizing that conversations remain private and do not influence AI outputs. This rollout, which started on February 9, 2026, represents a significant departure from CEO Sam Altman's May 2024 stance, where he described AI ads as "uniquely unsettling" and a "last resort." Now, with ChatGPT reporting 800 million weekly active users globally as of October 2025 and operational costs soaring, ads are being positioned as a necessary step to sustain growth without compromising higher paid tiers.

Efforts to monetize the platform have hit a snag in the past, but recent financial performance shows signs of recovery. Altman told employees in an internal memo that ChatGPT has returned to exceeding 10% monthly growth, and Codex, the company's coding tool, grew 50% in the past week. Without this ad revenue stream, the company might face tougher profitability pressures, especially as it scales to serve hundreds of millions of users. The test is limited to U.S. users for now, with no immediate plans for international expansion, though sources indicate that feedback from this beta will guide future adjustments.

Premium Advertising Terms and Industry Context

Advertisers participating in the test are required to commit a minimum of $200,000, with costs set at $60 CPM (cost per 1,000 views), a rate that rivals premium TV advertising like NFL games and far exceeds Meta's typical under-$20 CPM. This premium pricing reflects OpenAI's confidence in its massive user base and targeted reach, even as it forecasts 40 million new U.S. generative AI users by 2029. Industry analysts note that this move aligns with broader AI industry shifts toward hybrid ad-subscription models, where conversational platforms explore ads to fund expansion amid high operational expenses.

Privacy concerns linger, however. OpenAI has stated that ads avoid sensitive topics such as health, mental health, or politics, and exclude users predicted to be under 18. "We're focused on regulatory stability and user benefit," a spokesperson said in a brief statement, though attempts to reach Altman for further comment were unsuccessful. The rollout comes just after Anthropic aired a Super Bowl commercial on February 8 positioning itself as "no ads in AI," seen by many as a direct jab at OpenAI's strategy. Public reactions have been mixed, with some users criticizing the potential erosion of the ad-free experience, while advertisers eye the opportunity for targeted engagement in a growing market.

Short-Term Adjustments and Long-Term Implications

In the short term, OpenAI plans to adjust the beta based on user feedback, with an updated ChatGPT model launching this week and potential new ad formats in development. The company has teased positive internal metrics, suggesting sustained expansion, but experts caution that alienating users with ad aversion could backfire. "It's a delicate balance," one industry insider noted, "between monetizing scale and maintaining trust." Long-term, this test could reshape AI monetization, paving the way for global expansion and hybrid models, but risks include user backlash over privacy and the commodification of AI interactions.

Correction: An earlier version of this article misstated the start date of the ad test; it began on February 9, 2026, not February 8. OpenAI confirms that all ads are clearly labeled and dismissible, with no impact on AI outputs or data privacy.